ORIGINAL ARTICLE

A randomized, controlled trial of oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy Douglas K. Rex, MD,1 John McGowan, MPH,2 Mark vB. Cleveland, PhD,2 Jack A. Di Palma, MD3 Indianapolis, Indiana; Braintree, Massachusetts; Mobile, Alabama, USA

Background: No bowel preparation for colonoscopy is optimal with regard to efficacy, safety, and tolerability. New options for bowel preparation are needed. Objective: To compare a new hybrid preparation consisting of a reduced dose of oral sulfate solution (OSS) plus 2 L of sulfate-free electrolyte lavage solution (SF-ELS) with 2 low-volume preparations based on polyethylene glycol electrolyte lavage solution (PEG-ELS). Design: Two randomized, controlled trials. Setting: Twenty-four U.S. centers. Patients: A total of 737 outpatients undergoing colonoscopy. Interventions: In study 1, OSS plus SF-ELS was given as a split dose, and in study 2, OSS plus SF-ELS was given in its entirety the evening before colonoscopy. In study 1, the active control was 2 L of PEG-ELS plus ascorbic acid (PEG-EA) given as a split dose. In study 2, the control was 10 mg of bisacodyl plus 2 L of SF-ELS taken the evening before colonoscopy. Main Outcome Measurements: Rates of successful (good or excellent) bowel preparation. Results: In study 1, the rates of successful (excellent or good) preparation with OSS plus SF-ELS and PEG-EA were identical at 93.5% for split-dose preparation. OSS plus SF-ELS was noninferior to PEG-EA (P! .001). In study 2, OSS plus SF-ELS resulted in successful preparation in 89.8% of patients compared with 83.5% with bisacodyl plus SF-ELS in a same-day preparation regimen. OSS plus SF-ELS was noninferior to bisacodyl plus SF-ELS (P !.001). In study 1, vomiting was more frequent with OSS plus SF-ELS (13.5% vs 6.7%; P Z .042), and bloating was rated worse with PEG-EA (P Z .025). In study 2, overall discomfort was rated worse with OSS plus SF-ELS (mean score 2.1 vs 1.8; P Z .032). There were no deaths in either study and no serious adverse events considered related to the preparation. Limitations: Bowel cleansing was not scored by colon segment. Adenoma detection was not compared between the regimens. Conclusion: OSS plus SF-ELS is a new, safe, and effective bowel preparation for colonoscopy. (Gastrointest Endosc 2014;-:1-10.)

Abbreviations: ANOVA, analysis of variance; CMH, Cochran-MantelHaenszel; FDA, U.S. Food and Drug Administration; ITT, intent-to-treat; OSS, oral sulfate solution; PEG-EA, polyethylene glycol electrolyte lavage solution with ascorbic acid; PEG-ELS, polyethylene glycol electrolyte lavage solution; SF-ELS, sulfate-free electrolyte lavage solution. DISCLOSURE: The following authors disclosed financial relationships relevant to this article: Dr Rex is on the speakers’ bureau and has received research support from Braintree Laboratories, Inc. Dr Di Palma is a consultant medical director for, has received research support from, and is on the speakers’ bureau of Braintree Laboratories, Inc. Mr McGowan and Dr Cleveland are employees of Braintree Laboratories, Inc. All other authors disclosed no financial relationships relevant to this publication.

www.giejournal.org

Copyright ª 2014 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 http://dx.doi.org/10.1016/j.gie.2014.03.043 Received January 22, 2014. Accepted March 26, 2014. Current affiliations: Department of Medicine, Division of Gastroenterology, Indiana University Health, Indianapolis, Indiana (1), Braintree Laboratories, Inc., Braintree, Massachusetts (2), Division of Gastroenterology, University of South Alabama College of Medicine, Mobile, Alabama (3), USA. Reprint requests: Douglas K. Rex, MD, Indiana University Hospital, #4100, 550 N. University Boulevard, Indianapolis, IN 46202.

Volume

-,

No.

-

: 2014 GASTROINTESTINAL ENDOSCOPY 1

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

Bowel preparations are often considered the most problematic aspect of colonoscopy by patients.1-4 The tolerability of bowel preparations is related to their volume, taste, and side effects. Lower volume preparations with improved taste features have adequate efficacy for some patients and higher completion rates of ingestion and are better tolerated than large-volume preparations.5,6 Low-volume preparations are supplemented with a laxative such as bisacodyl7 or large-dose ascorbic acid.8 Bisacodyl, which rarely causes ischemic colitis, is included as a debulking agent before 2 L of sulfate-free electrolyte lavage solution (SF-ELS) (HalfLytely; Braintree Laboratories, Braintree, Mass). Recent trials showed that a hypertonic oral sulfate solution (OSS) (SUPREP; Braintree Laboratories) is an effective bowel preparation.1,2 In an internal evaluation, OSS purgative was found to be an adequate alternative to bisacodyl before low-volume SFELS ingestion. OSS followed by SF-ELS has been developed into a bowel preparation consisting of half of the labeled OSS bowel preparation dose followed by 2 L of SF-ELS (SUCLEAR; Braintree Laboratories). We report 2 randomized, controlled trials evaluating this hybrid OSS plus SF-ELS preparation as a split-dose regimen (study 1) with OSS taken the evening before colonoscopy followed by 2 L SF-ELS the following morning, or entirely as an evening-before regimen (study 2). For study 1 in which OSS was given the evening before colonoscopy and then 2 L SF-ELS the following morning (split dosing), we used polyethylene glycol electrolyte lavage solution with ascorbic acid (PEG-EA) (MoviPrep; Salix Pharmaceuticals, Morrisville, NC) as the active control because PEG-EA is approved by the U.S. Food and Drug Administration (FDA) for both evening-before and split dosing. For study 2, the active control was another FDA-approved bowel preparation intended for evening-before dosing (10 mg bisacodyl plus 2 L of SF-ELS; HalfLytely).

METHODS Study design These were single-blind, active-controlled, parallel studies of adult outpatients undergoing elective colonoscopy. The trials were registered at clinicaltrials.gov (study 1: NCT00756977, study 2: NCT00756548). Subjects were outpatients at least 18 years of age undergoing colonoscopy for routine indications, including screening, polyp surveillance, therapy, GI bleeding, evaluation of barium enema results, and change in bowel habits. Patients were excluded for known or suspected ileus, severe ulcerative colitis, GI obstruction, gastric retention, bowel perforation, toxic colitis, megacolon, impaired consciousness, indication of foreign-body removal or decompression, clinically significant electrolyte abnormalities on their precolonoscopy visit, previous significant GI surgeries (eg, colostomy, colectomy, gastric bypass), women who were pregnant or lactating or 2 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2014

Rex et al

Take-home Message
A combination of reduced dose of oral sulfate solution followed by 2 L of sulfate-free electrolyte lavage solution proved noninferior and safe compared with 2 lowvolume polyethylene glycol electrolyte lavage solution (PEG-ELS)–based preparations (2 L of PEG-ELS plus ascorbic acid and 10 mg bisacodyl plus 2 L of PEG-ELS).
The new preparation does not contain bisacodyl.

intended to become pregnant during the study interval, women of child-bearing potential who refused a pregnancy test, phenylketonuria, glucose-6-phosphate dehydrogenase deficiency, or those who were allergic to any preparation components. Patients could not have participated in an investigational surgical, drug, or device study within the previous 30 days. Laboratory testing was performed at a baseline precolonoscopy visit and the morning of their colonoscopy (after completing the preparation). Laboratory measurements included complete blood count, total protein, albumin, uric acid, alkaline phosphatase, alanine aminotransferase, amylase, aspartate aminotransferase, bicarbonate, blood urea nitrogen, calcium, chloride, creatine kinase, creatinine, direct bilirubin, g-glutamyl transferase, glucose, magnesium, serum osmolality, phosphate, potassium, sodium, total bilirubin, and total protein. A urine pregnancy test was performed on female patients of child-bearing potential.

Study centers Data were collected at 24 sites in the United States, all of which followed the same investigational protocols. Enrollment was competitive, and subjects were recruited in stand-alone gastroenterology and hospital-based practices. The experimental protocol and informed consent materials were approved by an institutional review board before study initiation. Written informed consent was obtained from all participating patients. Enrollment began on August 25, 2008, with the last patient enrolled on November 21, 2008.

Study medications The OSS purgative dose consisted of sodium sulfate 17.5 g, magnesium sulfate 1.6 g, and potassium sulfate 3.15 g and flavoring agents in aqueous liquid form supplied in a 6-oz plastic bottle. The content of the bottle was diluted with water to 16 oz for ingestion. The other element of the Suclear (Braintree Laboratories) preparation consists of 2 L of SF-ELS. The total preparation volume of OSS plus SF-ELS is 2480 mL (patients were also instructed to drink an additional 32 oz of water). In study 1, the active control PEG-EA was supplied in market packaging. The total preparation volume is 2000 mL, and patients are instructed to drink an extra 1000 mL of clear liquid. www.giejournal.org

Rex et al

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

In study 2, the control arm was bisacodyl and 2 L of SFELS, which was supplied in market packaging. All preparations were packaged in identical outer containers in an effort to maintain integrity of the treatment blind.

Study colon-cleansing methods In study 1, patients assigned to PEG-EA were instructed to eat a normal breakfast, light lunch, and clear soup and/ or plain yogurt for dinner. They were instructed to take clear liquids only from the time that the PEG-EA preparation was started until after completion of colonoscopy. In study 2, patients assigned to bisacodyl plus SF-ELS were instructed to consume only clear liquids on the day before colonoscopy. In study 1 (split dose), patients assigned to OSS plus SFELS were instructed to pour the contents of the 6-oz bottle of OSS into a provided mixing cup and to fill the cup with water to the fill line of 16 oz and then drink the entire cup of solution at about 6:00 PM. They were instructed to drink 1 additional 16-oz glass of water over the next 2 hours and then another 16-oz glass of water later in the evening. Patients were then instructed to begin drinking 2 L of SF-ELS (at a rate of about one 16-oz glass every 20 minutes) at 6:00 AM on the morning of colonoscopy. In study 2, patients who took their preparation the evening before the planned colonoscopy assigned to OSS plus SF-ELS were instructed to take the OSS at 6:00 PM in the same manner as in study 1. Then, 2 hours later (approximately 8:00 PM), patients were instructed to begin drinking the 2 L of SF-ELS at a rate one 16-oz glass every 20 minutes until the jug was empty. In both studies, patients were asked to return any portion of the SF-ELS preparation that they did not drink. The volume of the unused portion was recorded, and any amount greater than 4 oz was deemed noncompliance with ingestion. In study 1 (split dose), patients assigned to PEG-EA were instructed to pour the contents of pouches A and B into the 1-L container supplied with the commercial kit and fill it with water to the fill line. They were instructed to drink the solution over 1 hour at a rate of 8 oz every 15 minutes until complete and then to drink another 0.5 L of clear liquids that evening. This process was repeated starting at 6:00 AM on the morning of the colonoscopy. Compliance with PEG-EA ingestion was by patient report. In study 2 (evening-only dosing), patients assigned to bisacodyl and SF-ELS were instructed to take the two 5mg bisacodyl tablets with water between 12:00 PM and 3:00 PM on the day before colonoscopy. After waiting for a bowel movement (or a maximum of 6 hours after taking the bisacodyl), patients were instructed to drink the 2 L of SF-ELS solution at a rate of 8 oz every 10 minutes. Patients were instructed to return any portion of SF-ELS that they did not ingest, and any unused amount greater than 4 oz was considered noncompliance with ingestion. www.giejournal.org

For both study 1 and study 2, patients were randomly assigned to their respective bowel preparations in a 1:1 ratio at each participating site. The randomization schedule for this study was created by StatNet Statistical Services Network and was constructed by using random blocks of 2 balanced treatment assignments at each site. The randomization schedule was implemented by Braintree Laboratories before kit distribution to the site. After receipt of a sequential series of drug kits, site personnel dispensed the lowest numbered kit available to patients who met eligibility criteria to maintain the randomization schedule. To ensure that treatment blinding was maintained, the colonoscopist was not allowed to perform any drug-related activities, such as randomization, dispensing, return, or accountability. Patients were instructed not to discuss their study preparation with any staff member. The study was funded by Braintree Laboratories, and Braintree participated in the study design, study conduct, and data analysis.

Adequacy of bowel cleansing Bowel cleansing was scored by colonoscopists blinded to the preparation method. A 4-point scale used in previous bowel preparation trials1,2,5,7 was used, where 4 Z excellent (no more than small bits of adherent feces/fluid); 3 Z good (small amounts of feces or fluid not interfering with the examination); 2 Z fair (enough feces or fluid to prevent a completely reliable examination); 1 Z poor (large amounts of fecal residue, additional cleansing required). For the primary efficacy analysis, grades 3 and 4 were considered “successful” and grades 1 and 2 were considered “failures.” Each examination was also rated as to whether cleansing was adequate for examination and the need for repeat preparation.

Tolerance and safety measurements After completing the assigned preparation and before colonoscopy, patients in both studies completed a symptom questionnaire that asked them to provide an overall rating of their preparation-related symptoms of cramping, stomach bloating, nausea, and overall discomfort. Patients used a 5-point scale for each system, in which a score of 1 Z “none,” 2 Z “mild,” 3 Z “bothersome,” 4 Z “distressing,” and 5 Z “severely distressing.” This system for scoring symptoms caused by the preparation was used in previous studies.1,2,5,7 Safety assessments included adverse event monitoring as well as a physical examination performed at a prepreparation visit and a repeat physical examination performed after preparation (before colonoscopy). Blood samples for analysis were collected at the preprocedure visit and on the day of colonoscopy before the colonoscopy examination. Volume

-,

No.

-

: 2014 GASTROINTESTINAL ENDOSCOPY 3

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

Data analysis The primary efficacy analysis was based on an intent-totreat (ITT) analysis and included all patients randomized who took any portion of the study preparation. Patients who did not undergo colonoscopy because of inadequate preparation or preparation-related adverse events were considered failures for the primary efficacy analysis. For both studies, the success rate was analyzed by using the Cochran-Mantel-Haenszel (CMH) c2 test, adjusting for the effective investigator site. The formal hypothesis test result (P value) for treatment difference is presented together with a 2-sided 95% confidence interval for the difference. A lower confidence interval bound greater than 15% established noninferiority between the OSS plus SF-ELS and the control preparation in both studies. A 15% difference in preparation scores is considered to be clinically significant.8 For both study 1 and study 2, the protocol-planned study size was 360 patients. Patients were randomly assigned to 1 of the 2 preparations in a 1:1 ratio (180 patients per group). A dropout rate of approximately 5% per treatment group was expected. For study 1 (split dose), the efficacy of PEG-EA administered as a 2-day preparation was previously reported as 89%.6 Assuming a success rate of OSS plus SF-ELS of 85% or higher based on results of a similar oral sulfate–based 2-day preparation,1 a 2-sided 95% confidence interval (asymptotic Pearson c2 method) for the between-group success rates would result in a lower CI bound greater than 15%, establishing noninferiority between OSS plus SF-ELS and PEG-EA for a noninferiority margin of 15%. For study 2 (evening-only dosing), the efficacy of bisacodyl and SFELS administered as a 1-day preparation was previously reported as 87%.7 Assuming a success rate of OSS plus SF-ELS of 81% or higher based on results of a similar sulfate-based 1-day preparation,1 a 2-sided 95% confidence interval (asymptotic Pearson c2 method) for the between-group success rates will result in a lower CI bound greater than 15%, establishing noninferiority between OSS plus SF-ELS and bisacodyl and SF-ELS for a noninferiority margin of 15%. Secondary endpoints were analyzed in a fashion similar to the primary analysis by using a CMH c2 test adjusting for any side effects for counts (percentage), responses, and 2-way analysis of variance (ANOVA) with terms for treatment, site, and their interaction for mean responses. No adjustment was made for multiplicity testing of secondary endpoints. Results were presented for the effect results (P values) and 2-sided 95% confidence intervals for the treatment difference. All randomized patients were included in analyses of adverse events. Adverse events were summarized by sex, age subgroup (!65, R65, R75 years of age), and race by MedDRA category designations for body system and preferred term. The difference in adverse events between

4 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2014

Rex et al

study populations was tested by the c2 test or the Fisher exact test together with 95% confidence interval. Results of laboratory tests for the change from screening and group differences were tested by using ANOVA. In addition, shift tables are presented to describe changes in laboratory parameter values between baseline and posttreatment time points by using normal range categories (low, normal, high). A second analysis was performed by age subgroup. Symptom questionnaire data for patientreported individual symptoms for overall experience (nausea, stomach cramping, stomach bloating, and overall discomfort) were tested by using ANOVA with terms for treatment, site, and their interaction. The study was monitored by Premier Research (Quincy, Mass) and Braintree Laboratories. Funding was by Braintree Laboratories.

RESULTS Demographics The allocation of study subjects and their distribution in studies 1 and 2 is depicted in Figures 1 and 2. The cecum was intubated in 100% of patients in study 1 and 99% of patients in study 2. Table 1 shows the demographics for the study populations. Age, race, ethnicity, and weight were comparable in the study and active control groups in both studies; however, in study 1, more patients were female in the group receiving PEG-EA (58% vs 42%, P Z .017). The distribution of sexes in study 2 was comparable between study arms. In the 2 studies, a total of 780 patients were randomized, and 737 received a study preparation and were included in the intent-to-treat analysis. There were 188 patients 65 years of age or older, and 55 who were 75 years of age or older. Patients were asked to return any portions of their preparation not ingested. Compliant patients were defined as those who returned no more than 4 oz of solution or who reported taking all of the solution. In study 1, more PEG-EA patients completed the preparation compared with OSS plus SF-ELS patients (98% vs 90%, respectively; P Z .003). In study 2, more patients completed bisacodyl and SF-ELS compared with OSS plus SF-ELS (94% vs 87%, P Z .033). In study 2, in which the preparations were consumed on the day before colonoscopy, preparation time was significantly shorter with OSS plus SF-ELS, averaging 3.69 hours versus 5.52 hours for bisacodyl and SFELS (P ! .001).

Efficacy Table 2 shows the cleansing scores in the 2 studies. In the split-dose study (study 1), the success rate was 93.5% with OSS plus SF-ELS and 93.5% with PEG-EA, and no differences in the distribution of bowel prep

www.giejournal.org

Rex et al

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

Patients Screened N = 392 Screen Failures N=6 Did Not Meet Criteria = 5 Withdrew Consent = 1 Patients Randomized N = 386

OSS+SF-ELS N = 193

PEG-ELS-EA N = 193

ITT Patients N = 186

ITT Patients N = 185

ITT Completers N = 184

ITT Completers N = 185

ITT Non-completers N=2 Discontinuation Reason AE (nausea) = 1 Insurance issues = 1

ITT Non-completers N=0

Non-ITT Patients N=7 Withdrew Consent = 4 Lost to Follow-up = 2 Non-compliance = 1

Non-ITT Patients N=8 Withdrew Consent = 4 Non-compliance = 3 Lost to Follow-up = 1

Figure 1. Subject disposition in study 1. OSS þ SF–ELS, oral sulfate solution plus 2 L of sulfate-free electrolyte lavage solution; PEG-ELS þ EA, polyethylene glycol electrolyte lavage solution with ascorbic acid; ITT, intent-to-treat.

Patients Screened N = 399 Screen Failures N=5 Did Not Meet Criteria = 3 Withdrew Consent = 2 Patients Randomized N = 394

OSS+SF-ELS N = 196

Non-ITT Patients N = 20 Withdrew Consent = 14 Did Not Meet Criteria = 6

Bisacodyl+SF-ELS N = 198

ITT Patients N = 176

ITT Patients N = 190

ITT Completers N = 175

ITT Completers N = 187

ITT Non-completers N=1 Discontinuation Reason Adverse event = 1

ITT Non-completers N=3 Discontinuation Reason Non-compliance = 1 Lost to follow-up = 1 Non-study facility used = 1

Non-ITT Patients N=8 Withdrew Consent = 8

Figure 2. Subject disposition in study 2. OSS þ SF–ELS, oral sulfate solution plus sulfate-free electrolyte lavage solution; ITT, intent-to-treat.

www.giejournal.org

Volume

-,

No.

-

: 2014 GASTROINTESTINAL ENDOSCOPY 5

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

Rex et al

TABLE 1. Study demographics: intent-to-treat population Study 1: split dose OSS D SF-ELS

Study 2: evening only P value*

PEG-EA

OSS D SF-ELS

Bisacodyl D SF-ELS

Age, y

P value* .935

No. Mean (SD)

186

185

56.9 (11.4)

56.8 (11.0)

.903

176

190

56.8 (13.1)

56.9 (12.3)

Sex, no. (%)

1.000

Female

85 (46)

108 (58)

Male

101 (54)

77 (42)

White

160 (86)

159 (86)

Black

9 (5)

Other

.017

97 (55)

104 (55)

79 (45)

86 (45)

143 (81)

145 (76)

13 (7)

25 (14)

28 (15)

6 (3)

5 (3)

4 (2)

7 (4)

11 (6)

8 (4)

4 (2)

10 (5)

175 (94)

177 (96)

172 (98)

180 (95)

188 (44)

183 (46)

181 (42)

181 (42)

Racey .582

.662

Ethnicity Hispanic Non-Hispanic Weight, lb, mean (SD)

.639

.229

.176

.982

P values ! .05 are in bold. OSS þ SF-ELS, Oral sulfate solution plus 2 L of sulfate-free electrolyte lavage solution; PEG-EA, polyethylene glycol electrolyte lavage solution with ascorbic acid; Bisacodyl þ SF-ELS, 10 mg bisacodyl plus 2 L sulfate-free electrolyte lavage solution; SD, standard deviation. *P value from exact c2 test for the categorical variables and from an analysis of variance with term for treatment for the continuous variables. yPercentage for race does not equal 100% because Hispanic or Latino patients may not have reported a race.

TABLE 2. Bowel preparation scores Study 1: split dose

Study 2: evening only

OSSDSF-ELS, no. (%)

PEG-EA, no. (%)

P value*,y

OSSDSF-ELS, no. (%)

BisacodylDSF-ELS, no. (%)

P value*,y

4 (excellent)

96 (51.9)

95 (51.4)

.986

84 (47.7)

67 (35.6)

.010

3 (good)

77 (41.6)

78 (42.2)

74 (42.0)

90 (47.9)

2 (fair)

10 (5.4)

10 (5.4)

13 (7.4)

25 (13.3)

1 (poor)

1 (0.5)

2 (1.1)

4 (2.3)

5 (2.7)

Missing

1 (0.5)

0

1 (0.6)

1 (0.5)

3.46

3.44

.779

3.36

3.17

.016

181 (98)

180 (97)

.724

175 (97)

183 (98)

.744

Score

Mean scorez Adequate

P values ! .05 are in bold. OSSþSF-ELS, Oral sulfate solution plus 2 L of sulfate-free electrolyte lavage solution; PEG-EA, polyethylene glycol electrolyte lavage solution with ascorbic acid; BisacodylþSF-ELS, 10 mg bisacodyl plus 2 L of sulfate-free electrolyte lavage solution. *P value comparing excellent preparation is from a Cochran-Mantel-Haenszel test, controlling for site. yP value for mean score is from a 1-way analysis of variance with term for treatment. zPatient 26002 was excluded as nonassessable for efficacy analysis. Patient 25063 is included as “missing.”

cleansing scores or mean score were observed. In the evening-only dosing study (study 2), the success rate with OSS plus SF-ELS was 89.8% compared with 83.5% with bisacodyl and SF-ELS. OSS plus SF-ELS was 6 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2014

noninferior to both PEG-EA and bisacodyl and SF-ELS (P ! .001). However, OSS plus SF-ELS was associated with significantly more excellent preparations (47.7% vs 35.6%; P Z .01), and the average cleansing score was www.giejournal.org

Rex et al

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

TABLE 3. Mean patient symptom ratings at final visit Study 1: split dose OSSD SF-ELS (n [ 186)

PEG-EA (n [ 185)

Cramping (SD)

1.46 (0.71)

Stomach bloating (SD)

Study 2: evening only P valuey

OSSDSF-ELS (n [ 176)

BisacodylDSF-ELS (n [ 190)

P valuey

1.56 (0.76)

.330

1.50 (0.70)

1.55 (0.77)

.393

1.66 (0.73)

1.79 (0.76)

.025

1.74 (0.87)

1.62 (0.82)

.177

Nausea (SD)

1.73 (0.94)

1.54 (0.83)

.472

1.70 (1.02)

1.65 (0.97)

.818

Overall (SD)

1.87 (0.85)

1.90 (0.83)

.239

2.06 (0.98)

1.76 (0.80)

.032

Symptom*

P values ! .05 are in bold. OSSþSF-ELS, Oral sulfate solution plus 2 L sulfate-free electrolyte lavage solution; PEG-EA, polyethylene glycol electrolyte lavage solution with ascorbic acid; BisacodylþSF-ELS, 10 mg bisacodyl plus 2 L sulfate-free electrolyte lavage solution; SD, standard deviation. *Symptoms were scored as 1 Z none; 2 Z mild; 3 Z bothersome; 4 Z distressing; 5 Z severely distressing. yP value for difference between treatments by analysis of variance.

TABLE 4. Patients with treatment emergent adverse events* by MedDRA body system and preferred term in study 1 OSSDSF-ELS (n [ 193)

PEG-EA (n [ 147)

95% CIy

P value

155 (80.3)

147 (76.2)

4.1-12.4

.388

411

408

0

1 (0.5)

1.5 to 0.5

1.000

137 (71.0)

138 (71.5)

9.5 to 8.5

1.000

Abdominal distention

96 (49.7)

112 (58.0)

18.2 to 1.6

.126

Abdominal pain

68 (35.2)

77 (39.9)

14.3 to 5.0

.401

Upper abdominal pain

1 (0.5)

2 (1.0)

2.3 to 1.2

1.000

Abdominal tenderness

1 (0.5)

1 (0.5)

1.4 to 1.4

1.000

Dyspepsia

1 (0.5)

0

0.5 to 1.5

1.000

86 (44.6)

72 (37.3)

2.5 to 17.0

.178

Retching

0

2 (1.0)

2.5 to 0.4

.499

Vomiting

26 (13.5)

13 (6.7)

0.8 to 12.7

.042

116 (60.1)

121 (62.7)

 12.3 to 7.1

.676

Injuries (excoriation)

1 (0.5)

0

0.5 to 1.5

1.000

Investigations (laboratory)

5 (2.6)

2 (1.0)

1.1 to 4.2

.449

Musculoskeletal (muscle spasms)

1 (0.5)

0

0.5 to 1.5

1.000

Neoplasms (seborrheic keratosis)

0

1 (0.5)

1.5 to 0.5

1.000

Nervous system

3 (1.6)

2 (1.0)

1.7 to 2.8

1.000

Respiratory

1 (0.5)

0

0.5 to 1.5

1.000

Skin and tissue

2 (1.0)

1 (0.5)

1.2 to 2.3

1.000

Psychiatric (anxiety)

1 (0.5)

0

0.5 to 1.5

1.000

0

1 (0.5)

1.5 to 0.5

1.000

No. (%) of patients with any event No. of events Cardiac GI

Nausea

General disorders (discomfort)

Vascular (hypotension)

P values ! .05 are in bold. OSSþSF-ELS, oral sulfate solution plus 2 L of sulfate-free electrolyte lavage solution; PEG-EA, polyethylene glycol electrolyte lavage solution with ascorbic acid; CI, confidence interval. *Patients were counted once within each body system and preferred term. y95% confidence interval for difference in proportion and P value from the Fisher exact test.

www.giejournal.org

Volume

-,

No.

-

: 2014 GASTROINTESTINAL ENDOSCOPY 7

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

Rex et al

TABLE 5. Patients with treatment-emergent adverse events* by the MedDRA body system and preferred term in study 2 OSSDSF-ELS (n [ 196)

Bisacodyl DSF-ELS (n [ 198)

95% CIy

P value

145 (74.0)

142 (71.7)

6.5 to 11.0

.651

387

368

1 (0.5)

1 (0.5)

1.4 to 1.4

1.000

134 (68.4)

131 (66.2)

7.1 to 11.5

.668

Abdominal distention

92 (46.9)

85 (42.9)

5.8 to 13.8

.478

Abdominal pain

70 (35.7)

77 (38.9)

12.7 to 6.4

.533

Upper abdominal pain

1 (0.5)

1 (0.5)

1.4 to 1.4

1.000

Glossitis

1 (0.5)

0

0.5 to 1.5

.497

Hematemesis

1 (0.5)

0

0.5 to 1.5

.497

74 (37.8)

75 (37.9)

9.7 to 9.5

1.000

Retching

2 (1.0)

1 (0.5)

1.2 to 2.2

.622

Vomiting

18 (9.2)

15 (7.6)

3.9 to 7.1

.590

Projectile vomiting

1 (0.5)

0

0.5 to 1.5

.497

122 (62.2)

108 (54.5)

2.0 to 17.4

.127

3 (1.5)

3 (1.5)

2.4 to 2.4

1.000

0

1 (0.5)

1.5 to 0.5

1.000

1 (0.5)

1 (0.5)

1.4 to 1.4

1.000

No. (%) of patients with any event No. of events Cardiac GI

Nausea

General disorders (discomfort) Nervous system Respiratory Skin and tissue

OSSþSF-ELS, Oral sulfate solution plus 2 L of sulfate-free electrolyte lavage solution; BisacodylþSF-ELS, 10 mg bisacodyl plus 2 L of sulfate-free electrolyte lavage solution; CI, confidence interval. *Patients were counted once within each body system and preferred term. y95% confidence interval for difference in proportion and P value from the Fisher exact test.

higher with OSS plus SF-ELS than bisacodyl and SF-ELS (3.36 vs 3.17; P Z .016). Noninferiority testing revealed a highly statistically significant result (P ! .001), supporting the hypothesis that split-dose preparation with OSS plus SF-ELS is noninferior to split-dose PEG-EA and that evening-before dosing of OSS plus SF-ELS is noninferior to bisacodyl and SF-ELS. In study 1, the colonoscopists judged the preparation to be adequate in 98% of patients with OSS plus SF-ELS and in 97% of patients with PEG-EA. In study 2, 97% of patients had an adequate examination with OSS plus SF-ELS and 98% with bisacodyl and SF-ELS. In both studies, there was no difference between preparations in the elderly population (R65 years of age). In study 1, the percentage of successful preparations in the elderly population with OSS plus SF-ELS was 93% versus 86% with PEG-EA (P O .05). In study 2, both preparations had a success rate of 85% in the elderly population. In younger patients (!65 years), the success rate with OSS plus SF-ELS was 91%, which was significantly better compared with 83% observed with bisacodyl and SF-ELS 85% (P Z .03). 8 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2014

Table 3 lists ratings for abdominal symptoms associated with the preparations. Overall scores were low on a scale of 1 (none) to 5 (severely distressing). In study 1 (Table 4), there was a higher incidence of vomiting associated with OSS plus SF-ELS compared with PEG-EA (13.5% vs 6.7%; P Z .042). The higher incidence of vomiting with OSS plus SF-ELS was also detected in demographic subgroup analyses for white patients (P Z .023) and male patients (P Z .005). Patient symptom ratings for stomach bloating were higher for PEG-EA (P Z .025). In study 2, only patient-reported symptom ratings for overall discomfort were worse with OSS plus SF-ELS compared with bisacodyl and SF-ELS (mean symptom scores of 2.1 vs 1.8, P Z .032), although there was no difference in vomiting between study arms (9% with OSS plus SF-ELS vs 8% with bisacodyl and SF-ELS, P Z .590). Treatment-emergent adverse events were infrequent overall and similar between study groups (Table 5). One patient withdrew from study 1. This patient received OSS plus SF-ELS and experienced moderate nausea during consumption of the preparation and elected to withdraw. In study 2, 1 patient who received OSS plus www.giejournal.org

Rex et al

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

TABLE 6. Shifts from normal at baseline to outside the normal range on the day of colonoscopy Laboratory parameter (direction of change)

Study 1: split-dose regimen

Study 2: evening-only regimen

OSSDSF-ELS, % (n/N)

PEG-EA, % (n/N)*

OSSDSF-ELS % (n/N)*

Bisacodyl DSF-ELS % (n/N)*

Anion gap (high)y

10 (17/166)

8 (12/155)

3 (5/155)

5 (8/170)

Bicarbonate (low)

4 (6/170)

12 (20/161)

3 (4/156)

3 (5/167)

11 (19/170)

4 (6/162)

8 (12/157)

10 (17/170)

Calcium (high)

4 (6/141)

5 (7/144)

9 (12/139)

4 (5/139)

Chloride (low)

1 (1/173)

0 (0/163)

1 (1/157)

0 (0/171)

Creatinine (high)

1 (2/167)

1 (1/153)

2 (3/145)

3 (4/155)

Creatine kinase (high)

7 (10/147)

5 (7/143)

7 (10/138)

4 (6/151)

eGFR (low)z

9 (9/101)

11 (9/85)

25 (21/84)

14 (12/88)

Glucose (high)

2 (3/160)

3 (5/150)

7 (10/146)

3 (4/156)

Magnesium (low)

0 (0/169)

1 (1/163)

1 (1/158)

1 (1/169)

Osmolality (high)

4 (6/151)

8 (12/145)

2 (3/139)

5 (8/153)

Phosphate (low)

3 (5/171)

1 (2/160)

0 (0/155)

0 (0/168)

Potassium (low)

4 (6/162)

4 (7/159)

4 (5/144)

3 (4/160)

Sodium (low)

1 (1/169)

1 (1/163)

0 (0/157)

0 (0/169)

Uric acid (high)

4 (7/163)

3 (4/154)

6 (8/143)

7 (11/160)

Bilirubin, total (high)

OSSþSF-ELS, Oral sulfate solution plus 2 L of sulfate-free electrolyte lavage solution; PEG-EA, polyethylene glycol electrolyte lavage solution with ascorbic acid; BisacodylþSF-ELS, 10 mg bisacodyl plus 2 L of sulfate-free electrolyte lavage solution; eGFR, estimated glomerular filtration rate. *n/N, n Z number with the event and N Z number with normal baseline value. yAnion gap was calculated by using the following formula: [Sodium (mEq/L)]  [chloride (mEq/L)] þ [bicarbonate (mEq/L)]. Normal range was defined as 12  4, and high was defined as a calculated value greater than 16. zeGFR was calculated by using the Cockcroft-Gault formula.

SF-ELS withdrew after experiencing atrial fibrillation before colonoscopy and was discontinued by the investigator. The investigator determined that this event was unrelated to the study preparation. There were no deaths in either study. In study 1, 1 patient who received OSS plus SF-ELS underwent colonoscopy and was admitted to the hospital later that evening with complaints of severe abdominal pain. The patient had a fever, which resolved that day after antibiotic treatment. The patient was discharged after 2 days, with abdominal pain symptoms that resolved 4 days after colonoscopy. The investigator concluded that the serious adverse event was not related to bowel preparation. In study 2, there were no serious adverse events. Among the 186 patients 65 years of age and older in the 2 studies, there were no differences between treatment and control agents in treatment-emergent adverse events. There were 384 patients in the studies who had a medical history of cardiac or renal disease or hypertension or diabetes. Among these higher-risk patients, there were no differences in study 2 between OSS plus SF-ELS and bisacodyl plus SF-ELS in treatment-emergent adverse events. In study 1, the incidence of abdominal pain was 44.8% in www.giejournal.org

the PEG-EA arm and 30.4% in the OSS plus SF-ELS arm (P Z .041). No other differences were identified in study 1. There were no clinically significant changes in blood pressure, pulse, temperature, weight, or physical examination. In study 1, the mean pre- and postpreparation blood pressure increased by 4.84 mm Hg with OSS plus SF-ELS and 9.26 mm Hg with PEG-EA (P Z .014). There were no other statistically significant differences between OSS plus SF-ELS and the control preparations in either study for these parameters. Table 6 displays shifts in laboratory values from baseline to the postpreparation blood draw in the 2 studies. In both studies, there were some small differences between bowel preparations, but none of these differences were considered to be clinically significant.

DISCUSSION We describe 2 randomized, controlled trials testing the efficacy, safety, and tolerability of a bowel preparation composed of OSS followed by 2 L of SF-ELS. The preparation is similar to another FDA-approved low-volume Volume

-,

No.

-

: 2014 GASTROINTESTINAL ENDOSCOPY 9

Oral sulfate solution plus polyethylene glycol as a bowel preparation for colonoscopy

preparation for colonoscopy (bisacodyl and SF-ELS), but OSS is substituted for bisacodyl on the day before colonoscopy. The amount of OSS used is half the dose used in an FDA-approved preparation composed entirely of OSS (SUPREP; Braintree Laboratories). We found that OSS plus SF-ELS given in a split-dose regimen had a success rate (good or excellent) for bowel preparation in 93.5% of patients compared with 93.5% of patients who received split-dose PEG-EA. OSS solution plus SF-ELS had similar side effects compared with PEGEA, except that it resulted in slightly more patients with vomiting, likely attributed to the higher volume of preparation solution required on the morning of colonoscopy. Also, compliance (completion of ingestion of the preparation) was higher with PEG-EA than OSS plus SF-ELS. This might have resulted from the higher total volume of preparation fluid with OSS plus SF-ELS or perhaps because we measured compliance with PEG-EA only by patient report. OSS plus SF-ELS given the evening before resulted in successful preparation in 89.8% of patients compared with 83.5% of patients who received bisacodyl and SFELS. Excellent preparation scores and overall mean preparation score were both significantly higher with OSS plus SF-ELS, although lower than those observed for splitdose preparation, as reported in previous studies.1,2 Compliance (completion of ingestion of the preparation) was better with bisacodyl and SF-ELS, probably because the total prescribed volume of preparation was lower with bisacodyl and SF-ELS. Overall discomfort with OSS plus SF-ELS was slightly greater compared with bisacodyl and SF-ELS, possibly because of the greater preparation volume. In summary, these results show that OSS plus SF-ELS given in a split-dose regimen provides colon cleansing comparable to that with split-dose PEG-EA and when given as an evening-before dosing regimen provided better average cleansing scores compared with bisacodyl and SF-ELS. Tolerability as shown by side effects such as vomiting and overall discomfort was not improved compared with PEG-EA or bisacodyl and SF-ELS. The OSS plus

10 GASTROINTESTINAL ENDOSCOPY Volume

-,

No.

-

: 2014

Rex et al

SF-ELS preparation does not include bisacodyl and, therefore, may reduce the risk of ischemic colitis associated with bisacodyl. OSS plus SF-ELS provides an effective and safe bowel preparation regimen for colonoscopy. When possible, it should be administered as a split-dose preparation due to the higher cleansing efficacy observed with split-dosing in study 1 and the general superiority of split-dose bowel preparations reported in numerous published studies.9

REFERENCES 1. DiPalma JA, Rodriguez R, McGowan J, et al. A randomized clinical study evaluating the safety and efficacy of a new, reduced-volume, oral sulfate colon-cleansing preparation for colonoscopy. Am J Gastroenterol 2009;104:2275-84. 2. Rex DK, Di Palma JA, Rodriguez R, et al. A randomized clinical study comparing reduced-volume oral sulfate solution with standard 4-liter sulfate-free electrolyte lavage solution as preparation for colonoscopy. Gastrointest Endosc 2010;72:328-36. 3. Katz PO, Rex DK, Epstein M, et al. A dual-action, low-volume bowel cleanser administered the day before colonoscopy: results from the SEE CLEAR II study. Am J Gastroenterol 2013;108:401-9. 4. Rex DK, Katz PO, Bertiger G, et al. Split-dose administration of a dualaction, low-volume bowel cleanser for colonoscopy: the SEE CLEAR I study. Gastrointest Endosc 2013;78:132-41. 5. DiPalma JA, Wolff BG, Meagher A, et al. Comparison of reduced volume versus four liters sulfate-free electrolyte lavage solutions for colonoscopy colon cleansing. Am J Gastroenterol 2003;98:2187-91. 6. Ell C, Fischbach W, Bronisch HJ, et al. Randomized trial of low-volume PEG solution versus standard PEG þ electrolytes for bowel cleansing before colonoscopy. Am J Gastroenterol 2008;103:883-93. 7. DiPalma JA, McGowan J, Cleveland MV. Clinical trial: an efficacy evaluation of reduced bisacodyl given as part of a polyethylene glycol electrolyte solution preparation prior to colonoscopy. Aliment Pharmacol Ther 2007;26:1113-9. 8. Bitoun A, Ponchon T, Barthet M, et al. Results of a prospective randomised multicentre controlled trial comparing a new 2-L ascorbic acid plus polyethylene glycol and electrolyte solution vs. sodium phosphate solution in patients undergoing elective colonoscopy. Aliment Pharmacol Ther 2006;24:1631-42. 9. Rex DK, Johnson DA, Anderson JC, et al. American College of Gastroenterology guidelines for colorectal cancer screening 2008. Am J Gastroenterol 2009;104:739-50.

www.giejournal.org